10 Obesity Research Gaps
David Allison presented a thoughtful review this week of factors in research that interfere with understanding and reducing the impact of obesity. The occasion was the prestigious Atwater Memorial Lecture at the Experimental Biology meeting in San Diego.
While acknowledging considerable progress in nutrition and obesity, he provided a rundown of issues to address if we are to make greater progress. Here’s the list:
- Observational Studies. Too often, observational studies are used to support overreaching conclusions.
- Short-Term Studies. Likewise, short-term studies should serve as a cue for longer term research, not assumptions about long-term outcomes.
- Human Adaptation. Nutritional and behavioral interventions more often than not lead to human adaptation that can result in an effect being lost or even reversed.
- Repetition. Beyond a point, repeating observational studies or studies with weak designs serves only to create a bias of familiarity. A presumption repeated often enough may be accepted as proven.
- Rigorous Measurement. Self reports of dietary intake, weight, height, and physical activity are unreliable for scientific analysis. Yet a steady stream of publications emerges from analyses of self-reported data. They prove nothing but gain unwarranted attention.
- Control Groups. Nonspecific effects of control groups in trials of nutrition interventions can be exceedingly difficult to detect and eliminate.
- Conclusion Spinning. Numerous recent examples show that conclusions in both research publications and media reports are wont to conform to beliefs more than a strict interpretation of the data.
- Publication Bias. The relationship of breastfeeding to obesity is but one example of publication bias that leads to acceptance of an effect that rigorous analysis will not support.
- Inferring Causality. Unjustified use of causality language has been found at rates ranging from 15% to more than 50% in nutrition journals.
- Systematic Transparency. Systems such as clinical trial registries, CONSORT statements, and public data sharing will be essential for assuring the integrity of nutrition and obesity research.
Allison’s presentation tied back nicely to Wilbur Olin Atwater, for whom this lecture was named. Atwater struggled with politically mandated distortion of physiology textbooks to conform to the moralistic agenda of the Women’s Christian Temperance movement.
It’s worth asking if we’re vulnerable to such agendas in obesity.
Swiss Cheese Plant, photograph © grytr / flickr
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.